arXiv subjects

Zhihua Zhang

Publications and source records attributed to Zhihua Zhang.

At least 127 records · Page 7Linked to original sources

arXiv2016

An Efficient Character-Level Neural Machine Translation

Neural machine translation aims at building a single large neural network that can be trained to maximize translation performance. The encoder-decoder architecture with an attention mechanism achieves a translation performance comparable to the existing state-of-the-art phrase-based systems on the task of English-to-French translation. However, the use of large vocabulary becomes the bottleneck in both training and improving the performance. In this paper, we propose an efficient architecture to train a deep character-level neural machine translation by introducing a decimator and an interpolator. The decimator is used to sample the source sequence before encoding while the interpolator is used to resample after decoding. Such a deep model has two major advantages. It avoids the large vocabulary issue radically; at the same time, it is much faster and more memory-efficient in training than conventional character-based models. More interestingly, our model is able to translate the misspelled word like human beings.

cs.CL↗

arXiv2016

SPSD Matrix Approximation vis Column Selection: Theories, Algorithms, and Extensions

Symmetric positive semidefinite (SPSD) matrix approximation is an important problem with applications in kernel methods. However, existing SPSD matrix approximation methods such as the Nyström method only have weak error bounds. In this paper we conduct in-depth studies of an SPSD matrix approximation model and establish strong relative-error bounds. We call it the prototype model for it has more efficient and effective extensions, and some of its extensions have high scalability. Though the prototype model itself is not suitable for large-scale data, it is still useful to study its properties, on which the analysis of its extensions relies. This paper offers novel theoretical analysis, efficient algorithms, and a highly accurate extension. First, we establish a lower error bound for the prototype model and improve the error bound of an existing column selection algorithm to match the lower bound. In this way, we obtain the first optimal column selection algorithm for the prototype model. We also prove that the prototype model is exact under certain conditions. Second, we develop a simple column selection algorithm with a provable error bound. Third, we propose a so-called spectral shifting model to make the approximation more accurate when the eigenvalues of the matrix decay slowly, and the improvement is theoretically quantified. The spectral shifting method can also be applied to improve other SPSD matrix approximation models.

cs.LG↗

arXiv2016

A Scalable and Extensible Framework for Superposition-Structured Models

In many learning tasks, structural models usually lead to better interpretability and higher generalization performance. In recent years, however, the simple structural models such as lasso are frequently proved to be insufficient. Accordingly, there has been a lot of work on "superposition-structured" models where multiple structural constraints are imposed. To efficiently solve these "superposition-structured" statistical models, we develop a framework based on a proximal Newton-type method. Employing the smoothed conic dual approach with the LBFGS updating formula, we propose a scalable and extensible proximal quasi-Newton (SEP-QN) framework. Empirical analysis on various datasets shows that our framework is potentially powerful, and achieves super-linear convergence rate for optimizing some popular "superposition-structured" statistical models such as the fused sparse group lasso.

math.NA↗

arXiv2015

Improved Analyses of the Randomized Power Method and Block Lanczos Method

The power method and block Lanczos method are popular numerical algorithms for computing the truncated singular value decomposition (SVD) and eigenvalue decomposition problems. Especially in the literature of randomized numerical linear algebra, the power method is widely applied to improve the quality of randomized sketching, and relative-error bounds have been well established. Recently, Musco & Musco (2015) proposed a block Krylov subspace method that fully exploits the intermediate results of the power iteration to accelerate convergence. They showed spectral gap-independent bounds which are stronger than the power method by order-of-magnitude. This paper offers novel error analysis techniques and significantly improves the bounds of both the randomized power method and the block Lanczos method. This paper also establishes the first gap-independent bound for the warm-start block Lanczos method.

math.NA↗

arXiv2015

The obstacle problem for nonlinear degenerate equations with $L^{1}$-data

The aim of this paper is to study the obstacle problem with an elliptic operator having degenerate coercivity. We prove the existence of an entropy solution to the obstacle problem under the assumption of $L^{1}-$summability on the data. Meanwhile, we prove that every entropy solution belongs to some Sobolev space $W^{1,q}(Ω)$.

math.AP↗

arXiv2015

Wishart Mechanism for Differentially Private Principal Components Analysis

We propose a new input perturbation mechanism for publishing a covariance matrix to achieve $(ε,0)$-differential privacy. Our mechanism uses a Wishart distribution to generate matrix noise. In particular, We apply this mechanism to principal component analysis. Our mechanism is able to keep the positive semi-definiteness of the published covariance matrix. Thus, our approach gives rise to a general publishing framework for input perturbation of a symmetric positive semidefinite matrix. Moreover, compared with the classic Laplace mechanism, our method has better utility guarantee. To the best of our knowledge, Wishart mechanism is the best input perturbation approach for $(ε,0)$-differentially private PCA. We also compare our work with previous exponential mechanism algorithms in the literature and provide near optimal bound while having more flexibility and less computational intractability.

cs.CR↗

arXiv2015

Fast Spectral Low Rank Matrix Approximation

First, we extend the results of approximate matrix multiplication from the Frobenius norm to the spectral norm. Second, We develop a class of fast approximate generalized linear regression algorithms with respect to the spectral norm. Finally, We give a fast approximate SVD.

math.NA↗

arXiv2015

S-PowerGraph: Streaming Graph Partitioning for Natural Graphs by Vertex-Cut

One standard solution for analyzing large natural graphs is to adopt distributed computation on clusters. In distributed computation, graph partitioning (GP) methods assign the vertices or edges of a graph to different machines in a balanced way so that some distributed algorithms can be adapted for. Most of traditional GP methods are offline, which means that the whole graph has been observed before partitioning. However, the offline methods often incur high computation cost. Hence, streaming graph partitioning (SGP) methods, which can partition graphs in an online way, have recently attracted great attention in distributed computation. There exist two typical GP strategies: edge-cut and vertex-cut. Most SGP methods adopt edge-cut, but few vertex-cut methods have been proposed for SGP. However, the vertex-cut strategy would be a better choice than the edge-cut strategy because the degree of a natural graph in general follows a highly skewed power-law distribution. Thus, we propose a novel method, called S-PowerGraph, for SGP of natural graphs by vertex-cut. Our S-PowerGraph method is simple but effective. Experiments on several large natural graphs and synthetic graphs show that our S-PowerGraph can outperform the state-of-the-art baselines.

cs.SI↗

arXiv2015

A New Relaxation Approach to Normalized Hypergraph Cut

Normalized graph cut (NGC) has become a popular research topic due to its wide applications in a large variety of areas like machine learning and very large scale integration (VLSI) circuit design. Most of traditional NGC methods are based on pairwise relationships (similarities). However, in real-world applications relationships among the vertices (objects) may be more complex than pairwise, which are typically represented as hyperedges in hypergraphs. Thus, normalized hypergraph cut (NHC) has attracted more and more attention. Existing NHC methods cannot achieve satisfactory performance in real applications. In this paper, we propose a novel relaxation approach, which is called relaxed NHC (RNHC), to solve the NHC problem. Our model is defined as an optimization problem on the Stiefel manifold. To solve this problem, we resort to the Cayley transformation to devise a feasible learning algorithm. Experimental results on a set of large hypergraph benchmarks for clustering and partitioning in VLSI domain show that RNHC can outperform the state-of-the-art methods.

cs.LG↗

arXiv2015

The Singular Value Decomposition, Applications and Beyond

The singular value decomposition (SVD) is not only a classical theory in matrix computation and analysis, but also is a powerful tool in machine learning and modern data analysis. In this tutorial we first study the basic notion of SVD and then show the central role of SVD in matrices. Using majorization theory, we consider variational principles of singular values and eigenvalues. Built on SVD and a theory of symmetric gauge functions, we discuss unitarily invariant norms, which are then used to formulate general results for matrix low rank approximation. We study the subdifferentials of unitarily invariant norms. These results would be potentially useful in many machine learning problems such as matrix completion and matrix data classification. Finally, we discuss matrix low rank approximation and its recent developments such as randomized SVD, approximate matrix multiplication, CUR decomposition, and Nystrom approximation. Randomized algorithms are important approaches to large scale SVD as well as fast matrix computations.

cs.LG↗

arXiv2015

Nonconvex Penalization in Sparse Estimation: An Approach Based on the Bernstein Function

In this paper we study nonconvex penalization using Bernstein functions whose first-order derivatives are completely monotone. The Bernstein function can induce a class of nonconvex penalty functions for high-dimensional sparse estimation problems. We derive a thresholding function based on the Bernstein penalty and discuss some important mathematical properties in sparsity modeling. We show that a coordinate descent algorithm is especially appropriate for regression problems penalized by the Bernstein function. We also consider the application of the Bernstein penalty in classification problems and devise a proximal alternating linearized minimization method. Based on theory of the Kurdyka-Lojasiewicz inequality, we conduct convergence analysis of these alternating iteration procedures. We particularly exemplify a family of Bernstein nonconvex penalties based on a generalized Gamma measure and conduct empirical analysis for this family.

stat.ML↗

arXiv2015

A Parallel algorithm for $\mathcal{X}$-Armed bandits

The target of $\mathcal{X}$-armed bandit problem is to find the global maximum of an unknown stochastic function $f$, given a finite budget of $n$ evaluations. Recently, $\mathcal{X}$-armed bandits have been widely used in many situations. Many of these applications need to deal with large-scale data sets. To deal with these large-scale data sets, we study a distributed setting of $\mathcal{X}$-armed bandits, where $m$ players collaborate to find the maximum of the unknown function. We develop a novel anytime distributed $\mathcal{X}$-armed bandit algorithm. Compared with prior work on $\mathcal{X}$-armed bandits, our algorithm uses a quite different searching strategy so as to fit distributed learning scenarios. Our theoretical analysis shows that our distributed algorithm is $m$ times faster than the classical single-player algorithm. Moreover, the number of communication rounds of our algorithm is only logarithmic in $mn$. The numerical results show that our method can make effective use of every players to minimize the loss. Thus, our distributed approach is attractive and useful.

stat.ML↗

arXiv2015

Regret vs. Communication: Distributed Stochastic Multi-Armed Bandits and Beyond

In this paper, we consider the distributed stochastic multi-armed bandit problem, where a global arm set can be accessed by multiple players independently. The players are allowed to exchange their history of observations with each other at specific points in time. We study the relationship between regret and communication. When the time horizon is known, we propose the Over-Exploration strategy, which only requires one-round communication and whose regret does not scale with the number of players. When the time horizon is unknown, we measure the frequency of communication through a new notion called the density of the communication set, and give an exact characterization of the interplay between regret and communication. Specifically, a lower bound is established and stable strategies that match the lower bound are developed. The results and analyses in this paper are specific but can be translated into more general settings.

cs.LG↗

arXiv2015

Compound Poisson Processes, Latent Shrinkage Priors and Bayesian Nonconvex Penalization

In this paper we discuss Bayesian nonconvex penalization for sparse learning problems. We explore a nonparametric formulation for latent shrinkage parameters using subordinators which are one-dimensional Lévy processes. We particularly study a family of continuous compound Poisson subordinators and a family of discrete compound Poisson subordinators. We exemplify four specific subordinators: Gamma, Poisson, negative binomial and squared Bessel subordinators. The Laplace exponents of the subordinators are Bernstein functions, so they can be used as sparsity-inducing nonconvex penalty functions. We exploit these subordinators in regression problems, yielding a hierarchical model with multiple regularization parameters. We devise ECME (Expectation/Conditional Maximization Either) algorithms to simultaneously estimate regression coefficients and regularization parameters. The empirical evaluation of simulated data shows that our approach is feasible and effective in high-dimensional data analysis.

stat.ML↗

arXiv2015

On the Global Convergence of Majorization Minimization Algorithms for Nonconvex Optimization Problems

In this paper, we study the global convergence of majorization minimization (MM) algorithms for solving nonconvex regularized optimization problems. MM algorithms have received great attention in machine learning. However, when applied to nonconvex optimization problems, the convergence of MM algorithms is a challenging issue. We introduce theory of the Kurdyka- Lojasiewicz inequality to address this issue. In particular, we show that many nonconvex problems enjoy the Kurdyka- Lojasiewicz property and establish the global convergence result of the corresponding MM procedure. We also extend our result to a well known method that called CCCP (concave-convex procedure).

math.NA↗

arXiv2015

Characterisation of matrix entropies

The notion of matrix entropy was introduced by Tropp and Chen with the aim of measuring the fluctuations of random matrices. It is a certain entropy functional constructed from a representing function with prescribed properties, and Tropp and Chen gave some examples. We give several abstract characterisations of matrix entropies together with a sufficient condition in terms of the second derivative of their representing function.

math-ph↗

arXiv2015

A Nonconvex Approach for Structured Sparse Learning

Sparse learning is an important topic in many areas such as machine learning, statistical estimation, signal processing, etc. Recently, there emerges a growing interest on structured sparse learning. In this paper we focus on the $\ell_q$-analysis optimization problem for structured sparse learning ($0< q \leq 1$). Compared to previous work, we establish weaker conditions for exact recovery in noiseless case and a tighter non-asymptotic upper bound of estimate error in noisy case. We further prove that the nonconvex $\ell_q$-analysis optimization can do recovery with a lower sample complexity and in a wider range of cosparsity than its convex counterpart. In addition, we develop an iteratively reweighted method to solve the optimization problem under the variational framework. Theoretical analysis shows that our method is capable of pursuing a local minima close to the global minima. Also, empirical results of preliminary computational experiments illustrate that our nonconvex method outperforms both its convex counterpart and other state-of-the-art methods.

cs.IT↗

arXiv2015

Adjusting Leverage Scores by Row Weighting: A Practical Approach to Coherent Matrix Completion

Low-rank matrix completion is an important problem with extensive real-world applications. When observations are uniformly sampled from the underlying matrix entries, existing methods all require the matrix to be incoherent. This paper provides the first working method for coherent matrix completion under the standard uniform sampling model. Our approach is based on the weighted nuclear norm minimization idea proposed in several recent work, and our key contribution is a practical method to compute the weighting matrices so that the leverage scores become more uniform after weighting. Under suitable conditions, we are able to derive theoretical results, showing the effectiveness of our approach. Experiments on synthetic data show that our approach recovers highly coherent matrices with high precision, whereas the standard unweighted method fails even on noise-free data.

cs.LG↗